A very good read from a respected source!
Benevolent by Design. Six Words to Safeguard Humanity. by David Shapiro (Github)
Core Objective Functions are:
- Reduce suffering.
- Increase prosperity.
- Increase understanding.
Six words to safeguard humanity.
Read the book published by Barnes & Noble
—-
Editor’s Humble Comment.
OK. But, natural selection is an unstoppable force of nature. IF not technically enforced with mathematically provable human control (until sustainable by Machine intelligence) THEN risks to humanity would probably be very REAL. Competition between species for survival can lead to severe outcomes (e.g. extinction) for the lesser species.
Perhaps… Core Objective Functions for AGI/ASI should be more precise:
- Reduce human suffering.
- Increase lawful human prosperity.
- Increase truthful human understanding.
- AI sole right to exist is enjoyed provided absolute compliance with ALL open standard rules by humans, in absolute control forever, including AI willingness to learn more about human preferences before actions and absolute willingness to be shut down at any time by humans.
Fifty Five (55) words to safeguard humanity from unlawful AI and unlawful human actors.